Seismic Methods in Explor97 Master Page G Author Index Mineral Exploration O Explor97 Contents T Section Contents Paper 51 Previous Paper O Next Paper Application of 3-D Seismics to Mine Planning at Vaal Reefs Gold Mine, Number 10 Shaft, Republic of South Africa Pretorius, C.C.[1], Trewick, W.F.[1], and Irons, C.[2] 1. Anglo American Corp. of South Africa Ltd., Marshalltown, South Africa 2. Irons Geophysical Consulting, Henley-on-Thames, Oxfordshire, U.K. ABSTRACT During 1994, a three-dimensional (3-D) seismic reflection survey was undertaken at Vaal Reefs No. 10 Shaft with the objective of mapping the detailed structure of the reflective Ventersdorp Contact ‘Reef’ (VCR) gold ore body. This would provide vital input into future mine planning and development. The survey benefited from 10 years of two-dimensional seismic experience and one previous 3-D mine survey, conducted in the Witwatersrand Basin by the Gold Division of Anglo American Corporation of South Africa (AAC). The seismic survey at 10 Shaft has accurately and spectacularly delineated the 3-D structure of the VCR at depths ranging from 1000 m to 3500 m, imaging faults with throws in the 20 m to 1200 m range. The resultant structure plans have been satisfactorily validated by subsequent surface drilling and underground mining operations during the period 1994 to 1996. These plans have been merged with drill hole, underground survey and sampling data into an integrated mine modelling, gold reserve estimation and mine scheduling package. The Geology Department now manages the planning function at 10 Shaft and 3-D seismics has played a significant role in placing this important responsibility firmly within the geologists’ domain. Building on the success of the 10 Shaft survey, two further 3-D seismic surveys have been concluded over AAC group mines during 1996 and 1997. INTRODUCTION results have encouraged two more AAC mines to conduct 3-D seismic surveys in 1996–97, with the latest survey, at Western Ultra Deep Levels, The Gold Division of the Anglo American Corporation of South Africa being eight times the size of the motivational 10 Shaft survey and prob- (AAC) has successfully employed seismic reflection techniques in its ably one of the largest mine geophysical surveys undertaken to date in Witwatersrand Basin analysis programs since 1983. The emphasis dur- the mineral industry. ing the first ten years was on reconnaissance two-dimensional (2-D) This paper summarizes the 3-D seismic data acquisition, processing seismic surveys for subsurface structural mapping of the Witwatersrand and interpretation methodologies which were developed at 10 Shaft, and Triad rocks, particularly the auriferous Central Rand Group, within the will hopefully illustrate how this geophysical technique can make a pow- main Witwatersrand Basin (Figure 1). Building on the success of these erful contribution to the optimization of ore body extraction in the 2-D surveys, AAC’s first three-dimensional (3-D) seismic survey for dynamic Witwatersrand mining environment. mine planning and development took place at Western Deep Levels Gold Mine in 1993, with the second following a year later at Vaal Reefs Number 10 Shaft (Figure 1). GEOLOGICAL SETTING AND ITS RELATIONSHIP In many respects the 10 Shaft survey represents the mature applica- TO SEISMIC STRATIGRAPHY tion of seismics to detailed structural mapping in a deep AAC gold mine. Seismic structural maps of the Ventersdorp Contact Reef (VCR) gold Descriptions of Witwatersrand geology and associated seismic stratig- ore body at 10 Shaft have been satisfactorily validated by subsequent raphy appear in Pretorius et al. (1987, 1994), de Wet and Hall (1994) and drilling and mining operations between 1994 and 1996. These positive Weder (1994). Most of the gold in the Main Witwatersrand Basin occurs In “Proceedings of Exploration 97: Fourth Decennial International Conference on Mineral Exploration” edited by A.G. Gubins, 1997, p. 399–408 400 Seismic Methods in Mineral Exploration Figure 1: Regional location, surface and sub-surface geology of the Witswatersrand Basin (after Pretorious et al. (1986) with modification). Figure 2: Lithostratigraphic columns in the Witswatersrand Basin. Figure 3: (a) Generalized geological section No. 10 shaft. (b) Idealized west-east section across the No. 10 Shaft Lease Area showing unconformable relationships. (Ref: Trewick (1994) APPLICATION OF 3-D SEISMICS TO MINE PLANNING AT Pretorius, C.C., Trewick, W.F., and Irons C. VAAL REEFS GOLD MINE, NUMBER 10 SHAFT, REPUBLIC OF SOUTH AFRICA 401 Figure 4: A portion of 2-D seismic line OG-54 in the vicinity of 10 Shaft. in thin auriferous conglomerates, anomalously termed “reefs”, within into three subfacies. There is a very strong correlation between facies the predominantly arenaceous Central Rand Group (Figure 2). The type and gold grade as described by Trewick (1994). principal reef mined at 10 Shaft is the VCR. The VCR is found at the base Figure 4 shows part of a previous NW-SE 2-D seismic section tra- of the Klipriviersberg Group, which is the basal group of the Venters- versing the 10 Shaft 3-D survey area and clearly illustrates the seismic dorp Supergroup. The VCR is conformable with the overlying lavas but stratigraphy of the Ventersdorp and Witwatersrand Supergroups at this unconformable on the underlying quartzites of the Witwatersrand locality. Note especially the strong angular unconformity (event 2) Supergroup. Figure 3 is a generalized geological section through the where Platberg Group sediments are draped over underlying, tilted fault 10 Shaft Lease area, showing the unconformable relationship of the var- blocks of Klipriviersberg Group lavas. The drop in both P-wave velocity ious lithostratigraphic units. (from 6300 m/s to 5800 m/s) and density (from 2.9 g/cm3 to The VCR is a highly channelised reef with thicknesses varying from 2.67 g/cm3) as seismic waves pass from the Klipriviersberg lavas into the 20 cm to 400 cm. It can be divided into two major reef types, viz. plateau underlying Central Rand Group quartzites, produces a strong reflection reef and channel reef, with the channel reef being further subdivided coefficient, which fortuitously coincides with the VCR. Reflections at 402 Seismic Methods in Mineral Exploration Figure 5: 3-D seismic coverage at Vaal Reefs No. 10 Shaft. this lava contact (event 3) can therefore be used to map the VCR ore Note how a large, pre-Platberg normal fault such as f1-f1' (2000 m body. The Central Rand Group between events 3 and 4 maintains a seis- throw) on Figure 4 is imaged not only where it displaces the VCR, but mically transparent character in this portion of the Witwatersrand also where it traverses the reflective West Rand Group. However it is not Basin. The numerous positive and negative reflections in the Jeppestown easy to pick the exact position of this fault throughout the section, as cer- Subgroup of the West Rand Group below event 4 are caused by alternat- tain West Rand Group reflectors still appear to pass through it undis- ing shales and quartzites. This ‘stripy’ reflectivity characterizes most of turbed. This phenomenon is due to out-of-plane events being imaged the West Rand Group except for the Bonanza formation quartzites of the on this section, and it illustrates a major drawback of 2-D seismics, Government Subgroup, between events 5 and 6. These understandably which prevents its use as a detailed structural mapping tool. Another have a similar seismic signature to the Central Rand Group quartzites. excellent example of such ‘sideswipe’ is the steeply dipping fault plane reflector, labeled “fpr”, just above the interpreted position of f1-f1'. This reflector appears to originate from a strike extension of f1-f1', several Table 1: Vaal Reefs 10 Shaft 3-D seismic survey hundred metres north of the 2-D section line. data acquisition parameters. In order to map structure at the minimum resolution (20 m) expected by the mining clients and potentially available from the seismic Field crew Geoseis data at the VCR level, it is crucial to record 3-D seismic data, followed by Instrument type SN368 / CS2502 full 3-D depth migration at the processing phase. This will allow for the No. of channels 240 restoration of reflectors such as “fpr” to their true subsurface position in Record length 3 seconds 3-D space. Further discussions on the need for 3-D seismic imaging in Sample rate 2 milliseconds a structurally complex hard-rock mineral exploration environment Fold 2000% appear in Milkereit and Eaton (1996). Nominal Vibroseis Source Parameters Vibrator type Mertz M18 with Pelton Mk II electronics Pattern 4 Vibs in-line 62° W of N SEISMIC DATA ACQUISITION VP interval 40 m Array length 30 m Three-dimensional seismic survey design criteria must address issues Sweep length 16 seconds such as the subsurface area to be imaged; the required spatial resolution; Sweep frequency 10-90 Hz bin dimensions; fold of cover and the required source and receiver con- Gain 6 dB/octave boost figurations to achieve this; migration apertures and static control. Taper 0.3 sec Ashton et al. (1994) have published an interesting article on 3-D seismic Nominal Receiver Array Parameters survey design, including oil industry examples. Notes on appropriate modifications to seismic survey design criteria when addressing the Geophone type GCR hard-rock Witwatersrand environment appear in Pretorius et al. (1987). Geophone frequency 10 Hz From previous 2-D seismic surveys undertaken at 10 Shaft it was Station interval 40 m Pattern in-line clear that the peak frequencies achievable at the VCR would be between Spread 4 lines of 60 receivers 60 and 75Hz. Assuming an average velocity of 6000 m/s and quarter- wavelength vertical resolution criteria it was decided that 20 m vertical APPLICATION OF 3-D SEISMICS TO MINE PLANNING AT Pretorius, C.C., Trewick, W.F., and Irons C.